The use of amyloid-PET in memory clinic patients: AMYPAD Diagnostic and Patient Management Study

Abstract Background Amyloid‐PET allows the direct assessment of amyloid deposition, one of the main hallmarks of Alzheimer’s disease. However, this technique is currently not or incompletely reimbursed due to lack of randomized controlled studies demonstrating a clinical impact. Method AMYPAD‐DPMS is a prospective, multicenter, randomized controlled study. Patients with subjective cognitive decline plus (SCD+), mild cognitive impairment (MCI) or dementia from 8 European memory clinics were randomized into three study arms: ARM1, amyloid‐PET performed early in the diagnostic workup (within 1 month); ARM2, late in the diagnostic workup (after 8±2 months); or ARM3, if and when the managing physician chose to. The primary endpoint was the proportional difference in ARM1 and ARM2 participants in receiving an etiological diagnosis with very high diagnostic confidence (i.e. ≥ 90%) within 3 months. Secondary endpoints included changes in diagnosis and treatment plan. Result Participants were recruited from April 16th, 2018, to October 30th, 2020. There were 272 participants in ARM1, 260 ARM2, and 261 ARM3 that underwent both baseline and 3‐month visit. 88% of ARM3 participants underwent amyloid‐PET within 3 months, and the average time from baseline to prescribe amyloid‐PET was 46 (IQR=58) days. After 3 months, 40% (109/272) of ARM1 participants and 37% (97/261) of ARM3 had a diagnosis with very high confidence vs 11% (30/260) in ARM2 (p<0.001). This was consistent across clinical stages (SCD+: 30%, 22%, and 6%, p<0.05; MCI: 42%, 39%, and 9%, p<0.05; dementia: 49%, 49%, and 20%, p<0.05). Changes in diagnosis were more frequent in ARM1 (44%) versus ARM3 (29%, p=0.001) and ARM2 (11%, p<0.001), and in ARM3 versus ARM2 (p<0.001). In participants for whom an etiological diagnosis of AD or non‐AD was confirmed after 3 months, changes in diagnostic confidence were greater in ARM1 (AD: +14%; non‐AD: +12%) and ARM3 (+11%; +10%) vs ARM2 (+1%, +1%; p<0.05). Conclusion An amyloid‐PET performed in the early phases of a diagnostic workup is associated with a greater proportion of etiological diagnoses with very high confidence and more frequent changes in diagnosis and higher diagnostic confidence after 3 months. This evidence supports the implementation of this technique early in the diagnostic workup.